I’ve just posted the following draft:

> A new version of I-D, draft-bellis-dnsop-connection-close-00.txt
> has been successfully submitted by Ray Bellis and posted to the
> IETF repository.
> 
> Name:         draft-bellis-dnsop-connection-close
> Revision:     00
> Title:                Connection Close Signalling for DNS
> Document date:        2014-10-24
> Group:                Individual Submission
> Pages:                5
> URL:            
> http://www.ietf.org/internet-drafts/draft-bellis-dnsop-connection-close-00.txt
> Status:         
> https://datatracker.ietf.org/doc/draft-bellis-dnsop-connection-close/
> Htmlized:       
> http://tools.ietf.org/html/draft-bellis-dnsop-connection-close-00
> 
> 
> Abstract:
>   This document updates [RFC6891] by specifying a new single-bit flag
>   in a DNS response that when seen in a packet carried over a
>   connection-orientated transport protocol indicates to the client that
>   it should close the current connection.

It’s intended as a simple mechanism to improve TCP connection handling that’s 
far closer to HTTP’s “Connection: close” semantics than that proposed in 
draft-wouters-edns-tcp-keepalive.

There are several outstanding questions therein - some relating to TCP state 
where I don’t know the technical answer, and a couple where opinions from the 
WG would be welcomed.  The latter mostly relate to whether it’s useful to 
define a request-side semantic for the proposed flag in addition to the 
response-side semantics already written.

kind regards,

Ray

_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to